Site-Directed Chemical Cross-Linking Demonstrates that Helix IV Is Close toHelices VII and XI in the Lactose Permease†

Jianhua Wu,‡ Dorothy Hardy,§ and H. Ronald Kaback*,§

Department of Physiology, Howard Hughes Medical Institute, Departments of Physiology and of Microbiology and MolecularGenetics, Molecular Biology Institute, UniVersity of California, Los Angeles, Los Angeles, California 90095-1662

ReceiVed September 30, 1998

ABSTRACT: The N-terminal six transmenbrane helices (N6) and the C-terminal six transmembrane helices(C6) of the lactose permease, each containing a single-Cys residue, were coexpressed, and proximity wasstudied. Paired Cys residues in helices IV (positions 114, 116, 119, 122, 125, or 129) and VII (227, 231,232, 234, 235, 238, 239, 242, 243, 245, or 246) or XI (350, 353, 354, 357, 361, or 364) were tested forcross-linking in the presence of two rigid homobifunctional thiol-specific cross-linkers,N,N′-o-phenylenedimaleimide (o-PDM; 6 Å) andN,N′-p-phenylenedimaleimide (p-PDM; 10 Å). Cys residues inthe middle of helix IV (position 119 or 122) cross-link to Cys residues in the middle of helix VII (position238, 239, 242, or 243). In contrast, no cross-linking is evident with paired Cys residues at either end ofhelix IV (position 114, 116, 125, or 129) or helix VII (position 227, 231, 232, 234, 235, 245, or 246). Onthe other hand, Cys residues in the cytoplasmic half of helix IV (position 125 or 129) cross-link with Cysresidues in the cytoplasmic half of helix XI (position 350, 353, or 354), while paired Cys residues at theperiplasmic ends of the two helices do not cross-link. The results indicate that helices IV and VII crossin a scissors-like manner with the cytoplasmic end of helix IV tilting toward helix XI.

The lactose permease (lac permease)1 of Escherichia coliis a paradigm for secondary transport proteins that couplefree energy stored in electrochemical ion gradients into asolute concentration gradient (reviewed in refs1-5). Thishydrophobic, polytopic membrane protein which catalyzesthe coupled stoichiometric translocation ofâ-galactosides andH+ has been solubilized, purified to homogeneity, reconsti-tuted into proteoliposomes, and shown to be solely respon-sible for â-galactoside transport (reviewed in ref6) as amonomer (see ref7). All available evidence (reviewed inrefs 8-10) indicates that the permease is composed of 12hydrophobic or amphipathicR-helical rods that traverse themembrane in zigzag fashion connected by relatively hydro-philic loops with the N and C termini on the cytoplasmicface.

Site-directed mutagenesis and Cys-scanning mutagenesishave not only allowed delineation of amino acid residues inthe permease that are essential for active transport and/orsubstrate binding, but have also yielded a unique library ofmutants with a single-Cys residue at each position in thepermease (reviewed in ref11). With the single-Cys mutant

library, a variety of site-directed approaches which includeexcimer fluorescence, chemical cleavage, engineered divalentmetal binding sites, electron paramagnetic resonance spec-troscopy, thiol-specific cross-linking, and identification ofdiscontinuous mAb epitopes, has led to a general helixpacking model of lac permease (reviewed in refs9, 10, 12)(Figure 2). Although many of the proximity relationshipsbetween helices have been documented by more than oneexperimental approach (see ref10), helical tilting is particu-larly interesting since it has been shown (13-16) that asolvent-filled cleft or notich is present in the permease whichmay represent part of the substrate translocation pathway.Recently (17-19), tilting of helices I, II, V, VII, VIII, X,and XI has been demonstrated by site-directed disulfide orchemical cross-linking of paired Cys resides in a functionalpermease construct expressed in two nonoverlapping con-tiguous fragments (N6C6 permease; ref20).

In this communication, we examine the spatial relation-ships between helices IV and VII or XI. The resultsdemonstrate that helices IV and VII are in close proximityat about the middle of the two transmembrane domains, whilehelices IV and XI are in close proximity at the cytoplasmicends. Binding of ligand enhances cross-linking efficiencybetween certain pairs of Cys residues in helices IV and VII,but not between Cys residues in helices IV and XI,suggesting that the interface between helices IV and VII maybe conformationally active.

EXPERIMENTAL PROCEDURES

Materials. Protein A-conjugated horseradish peroxidase(PA-HRP) enhanced chemiluminescence (ECL) detection kitswere obtained from Amersham (Arlington Heights, IL).

† This work was supported in part by NIH Grant DK51131.* Corresponding author mailing address: HHMI/UCLA, 5-748

MacDonald Research Laboratories, Box 951662, Los Angeles, Cali-fornia 90095-1662. Tel: (310) 206-5053. Fax: (310) 206-8623.

‡ Department of Physiology.§ Howard Hughes Medical Institute.1 Abbreviations: lac, lactose; C-less permease, functional lac per-

mease devoid of Cys residues; N6, the N-terminal six transmembranehelices; C6, the C-terminal six transmembrane helices; TDG,â,D-galactopyranosyl 1-thio-â,D-galactopyranoside; IPTG, isopropyl 1-thio-â,D-galactopyranoside;o-PDM, N,N′-o-phenylenedimaleimide;p-PDM,N,N′-p-phenylenedimaleimide; NaDodSO4/PAGE, sodium dodecylsulfate/polyacrylamide gel electrophoresis; NEM,N-ethylmaleimide.

1715Biochemistry1999,38, 1715-1720

10.1021/bi982342b CCC: $18.00 © 1999 American Chemical SocietyPublished on Web 01/20/1999

Avidin-conjugated horseradish peroxidase (avidin-HRP) waspurchased from Pierce (Rockford, IL).N,N′-o-phenylenedi-maleimide (o-PDM) andN,N′-p-phenylenedimaleimide (p-PDM) were from Sigma (St. Louis, MO).

Construction of N6/C6 Permease with Single-Cys Residues.Construction of permease mutants containing single-Cysreplacements in helices IV, VII, and XI has been described(11, 21, 22). To each mutant with a single-Cys replacementat positions 114, 116, 119, 122, 125, or 129 in helix IV, thebiotin acceptor domain from theKlebsiella pneumoniaeoxaloacetate decarboxylase was inserted into the middlecytoplasmic loop as described (23). The 3′ half of the lacYgene in each construct was then deleted byAflII digestionfollowed by intramolecular ligation resulting in plasmid pN6

which encodes N6 with a single-Cys residue at a givenposition and the biotin acceptor domain at the C terminus(Figure 1). Construction of plasmid pC6 which encodes theC-terminal six transmembrane helices of lac permease hasbeen described (17, 24). To introduce a single-Cys residuein helix VII (positions 227, 231, 232, 234, 235, 238, 239,242, 243, 245, or 246) or in helix XI (positions 350, 354,357, 361, or 364) in C6, theBstXI-HindIII fragment of pC6

was replaced with the corresponding DNA fragment from aspecified single-Cys permease mutant (Figure 1). Each Cysreplacement mutant in N6 or C6 was verified by usingdideoxynucleotide termination (25).

Expression of N6/C6 Permease and Membrane Prepara-tion. E. coli HB101 (lacY-Z+) was transformed with bothpN6 and pC6, each encoding a given permease fragmentcontaining a specified single-Cys residue. Cultures (50 mL)were grown at 37°C in Luria-Bertani broth containing 100µg/mL ampicillin and 20µg/mL chloramphenicol to an OD600

of 1.0 and induced with 1 mM isopropyl 1-thio-â,D-galactopyranoside for 2 h. Cells were harvested by centrifu-gation and washed once with 20 mM Tris-HCl (pH 7.4)/2mM ethylenediaminetetraacetate and suspended in the samebuffer followed by incubation with 100µg/mL lysozyme for10 min in ice. Membranes were prepared by sonification,harvested by centrifugation, and suspended in 20 mM Tris-HCl (pH 7.4).

Chemical Cross-Linking and Analysis. All cross-linkingexperiments were carried out by addingo-PDM or p-PDM

(0.5 mM final concentration) to membrane preparations at aprotein concentration of 2 mg/mL. Reactions were carriedout at room temperature for 30 min and terminated by addingsodium dodecyl sulfate (NaDodSO4) sample buffer contain-ing 5% (v/v) â-mercaptoethanol at given times. Iodine-catalyzed disulfide cross-linking was carried out as described(17). Samples were subjected to electrophoresis in Na-DodSO4/12% polyacrylamide gels (NaDodSO4/PAGE). C6

was detected by immunoblotting with rabbit polyclonalantibody against a C-terminal dodecapeptide correspondingto the last twelve amino acid residues of the permease (26).N6 with a biotin acceptor domain at the C terminus wasdetected with avidin-conjugated horseradish peroxidase (27).Cross-linked N6/C6 reacts with both anti-C-terminal antibodyand avidin-HRP.

Protein Assays. Protein was assayed by using a MicroBCA protein determination kit (Pierce).

RESULTS

Cross-Linking N6/C6 Permease with Paired Cys Residuesin Helices IV and VII.Previous in situ thiol-specific cross-linking studies with N6/C6 permease (17-19, 28) demonstratethat the periplasmic halves of helices VII and I or II are inclose proximity. To examine whether helices IV and VII arein close proximity, we placed paired Cys residues at variouspositions along one face of helix IV (position 114, 116, 119,122, 125, or 129) in N6 and along one face of helix VII(position 227, 231, 232, 234, 235, 238, 239, 242, 243, 245,or 246) in C6 (Figures 1 and 2). N6 and C6 each with a Cysresidue at a given position were coexpressed, and proximitywas assessed by thiol-specific chemical cross-linking in situ.The homobifunctional thiol-specific cross-linkerso-PDM andp-PDM were chosen because of their relatively short lengthand hydrophobicity (28). Hydrophobicity is presumablyimportant because the Cys residues are thought to be in ahydrophobic environment within the membrane. In addition,o-PDM andp-PDM are rigid cross-linking agents in whichthe maleimido groups are coupled to benzene rings in theortho or para position at fixed distances of about 6 and 10Å, respectively. C6 which reacts with anti-C-terminal anti-body migrates with anMr of about 20 kDa (Figure 3A), N6with the biotin acceptor domain which reacts with avidin-

FIGURE 1: Secondary structure model of N6/C6 split permease. Lac permease is shown as the N-terminal six transmembrane helices (N6)and the C-terminal six transmembrane helices (C6). N6 has a biotin acceptor domain (BD) at the C terminus. Single-Cys replacements inhelix IV (positions 114, 116, 119, 122, 125, and 129), helix VII (positions 227, 231, 232, 234, 235, 238, 239, 242, 243, 245, and 246), andhelix XI (positions 350, 353, 354, 357, 361, and 364) are numbered and highlighted.

1716 Biochemistry, Vol. 38, No. 6, 1999 Wu et al.

HRP migrates with anMr of about 35 kDa (Figure 3B), andcross-linked N6/C6 which reacts with both anti-C-terminal

antibody and avidin-HRP migrates with anMr of about 52kDa (Figure 3A,B).

When membranes containing N6/C6 permease with pairedCys residues at positions in the periplasmic ends of helicesIV (position 114 or 116) and VII (position 245 or 246) aretreated witho-PDM or p-PDM, no cross-linking is detected(Table 1). Similarly, no cross-linking is observed when pairedCys residues are placed in the cytoplasmic ends of helicesIV (position 125 or 129) and VII (position 227, 231, 232,234, or 235) (Table 1). In marked contrast, however, whenpaired Cys residues are placed at positions in the middle oftransmembrane helices IV and VII, cross-linking is clearlyevident (Figure 3; Table 1). N6/C6 with Cys pairs 119/238,119/242, 119/243, 122/238, or 122/239 is readily cross-linkedby o-PDM or p-PDM, indicating close proximity at themiddle of the two transmembrane domains. Thus, helicesIV and VII appear to cross in a scissors-like manner.

Cross-Linking N6/C6 Permease with Paired Cys Residuesin Helices IV and XI. To study tilting of helix IV with respectto helix XI, we coexpressed N6 with a Cys residue at variouspositions in helix IV with C6 containing a Cys residue atposition 350, 353, 354, 357, 361, or 364 in helix XI (Figures1 and 2). When membranes containing N6/C6 with Cys pairs114/364, 116/364, 116/361, 119/361, 119/257, 122/361, or122/357 are exposed too-PDM or p-PDM, no cross-linkingis evident (Figure 4B and Table 1), indicating that the

FIGURE 2: Helix packing in lac permease. The six essential residues [Glu126 (helix IV), Arg144 (helix V), Glu269 (helix VIII), Arg302(helix IX), and His322 and Glu325 (helix X)] and two interacting pairs of Asp-Lys residues [Asp237 (helix VII)/Lys358 (helix XI) andAsp240 (helix VII)/Lys319 (helix X)] are highlighted. Positions of NEM-sensitive Cys replacements are indicated with a small black dot.Positions where single-Cys mutants are protected from alkylation by substrate (145, 148, 264, 268, and 272) are crosshatched. Cys replacementmutants in helices IV, VII, and XI tested for cross-linking in this study are shown as filled circles (positive cross-linking between helicesIV and VII), filled cicles with a white dot (positive cross-linking between helices IV and XI), or unfilled squares (negative cross-linking).

FIGURE 3: Cross-linking of N6/C6 permease with paired Cysresidues in helices IV and VII. Membranes were prepared fromcells expressing N6 and C6, each with a single-Cys residue at givenpositions as indicated. Cross-linking was carried out with 0.5 mMo-PDM or p-PDM as indicated at room temperature for 30 min.Reactions were terminated by adding NaDodSO4 sample buffercontaining 5% (v/v)â-mercaptoethanol. Samples containing ap-proximately 20µg of membrane protein were subjected to Na-DodSO4/PAGE and electroblotted. The blots were probed with anti-C-terminal antibody or avidin-HRP. The positions of N6, C6, andN6/C6 are indicated by arrows: (A) cross-linking of Cys pairs 119/242 probed with anti-C-terminal antibody; (B) cross-linking of Cyspairs 119/242 probed with avidin-HRP; (C) cross-linking of Cyspairs 119/243 probed with avidin-HRP; (D) cross-linking of Cyspairs 122/238 probed with avidin-HRP; and (E) cross-linking ofCys pairs 122/239 probed with avidin-HRP.

Helix Tilting in lac Permease Biochemistry, Vol. 38, No. 6, 19991717

periplasmic halves of helices IV and XI are not in closeproximity. However, when paired Cys residues are placedat the cytoplasmic ends of the two transmembrane domains,cross-linking is readily observed. Thus, N6/C6 expressing Cyspairs 125/354, 129/350, 125/353, or 129/353 is clearly cross-linked by o-PDM or p-PDM (Figure 4A and Table 1). Theresults show that helix IV is tilted in such a manner as tocross helix VII at its approximate middle with the cytoplas-mic end within 6-10 Å of the cytoplasmic end of helix XI.

Ligand-Induced Changes in Cross-Linking. Cross-linkingof N6/C6 in situ is a sensitive probe for conformationallyactive interfaces between transmembrane helices (18, 19, 28).Interhelical distance changes have been observed at theperiplasmic interface between helix VII and helix I or II.To test whether the interface between helices IV and VII orIV and XI is conformationally active, we carried out cross-linking of N6/C6 containing paired Cys residues in theabsence or presence of 10 mMâ,D-galactopyranosyl 1-thio-â,D-galactopyranoside (TDG). With N6/C6 expressing pairedCys residues at positions 119 and 238, cross-linking byo-PDM or p-PDM is weak in the absence of ligand (Figure5A). Interestingly, cross-linking becomes stronger with bothcross-linking agents in the presence of TDG (Figure 5A).

Similar changes are also evident in N6/C6 expressing pairedCys residues at positions 119/242 (Figure 5B). Since cross-linking efficiency with the 6 or the 10 Å cross-linking agentis increased by TDG, one possibility is that binding of TDGincreases the reactivity of the Cys residue at position 119.Consistently, when a Cys residue at position 122 is pairedwith a Cys residue at position 238 or 242, cross-linkingefficiency does not change in the presence of TDG (datanot shown). Moreover, TDG binding does not alter iodine-catalyzed disulfide cross-linking of Cys pairs 119/238 or 119/242 (data not shown), indicating that inter-thiol distance isnot changed. Finally, binding of TDG does not alter cross-linking efficiency with N6/C6 containing Cys pairs 125/354or 129/350 (data not shown), suggesting that the interfacebetween helices IV and XI is not conformationally active.

DISCUSSION

Site-directed mutagenesis of lac permease has allowed theidentification of 6 residues that are irreplaceable with respectto substrate binding and active transport: Glu126 (helix IV)and Arg144 (helix V) which are indispensable for substratebinding and recognition; and Glu269 (helix VIII), Arg302(helix IX), His322, and Glu325 (helix X) which are essentialfor H+ translocation and coupling (reviewed in refs10, 12).However, structural information is a prerequisite for under-standing the mechanism of the transport mechanism at themolecular level (see ref12). In this regard, a detailed helixpacking model which includes the tilts of the helices iscritical for delineation of the spatial relationship betweenthe irreplaceable residues and to define the substrate trans-location pathway. Although the permease has been impervi-ous to crystallization attempts which precludes high-resolution structure analysis, we have utilized a wide varietyof site-directed biophysical and biochemical techniques toobtain relatively high-resolution structural information re-garding lac permease. In particular, use of site-directed thiolcross-linking with split permease in situ has facilitated thedetermination of helix proximities, tilting, and ligand-inducedinterhelical distance changes (17-19, 28-30). Cross-linkingstudies with paired Cys residues in two helices across thediscontinuity in N6C6 permease have provided proximity andtilting data between helices I and VII, II and VII, II and XI,V and VII, V and VIII, and V and X (18, 19). In this paper,the tertiary contacts of helix IV with helices VII and XI inthe C-terminal half of the permease are documented. As

Table 1: Chemical Cross-Linking of Paired Cys Residues inHelices IV/VII or IV/XI a

IV/VIIo-PDM(6 Å)

p-PDM(10 Å) IV/XI

o-PDM(6 Å)

p-PDM(10 Å)

129/227 - - 129/350 +++ +++125/231 - - 129/353 + ++125/232 - - 125/353 + ++125/234 - - 125/354 +++ +++125/235 - -122/238 +++ +++ 122/357 - -122/239 +++ +++ 122/361 - -119/238 +++ +++ 119/357 - -119/242 +++ +++ 119/361 - -119/243 ++ +116/245 - - 116/361 - -116/246 - - 116/364 - -113/245 - -114/246 - - 114/364 - -

a Crosslinking experiments were carried out at 22°C for 30 min asdescribed for Figure 3 and probed with avidin-HRP:+++, >30%efficiency; ++, 10-30% efficiency; +, <10% efficiency; -, nodetectable crosslinking.

FIGURE 4: Cross-linking of N6/C6 permease with paired Cysresidues in helices IV and XI. Membranes were prepared from cellsexpressing N6 and C6, each with a Cys residue at the positionindicated. Cross-linking was carried out at room temperature byincubating membranes (2 mg of protein/mL) with 0.5 mMo-PDMor p-PDM as indicated for 30 min. Reactions were terminated byadding NaDodSO4 sample buffer containing 5% (v/v)â-mercap-toethanol. Samples were analyzed as described in Figure 3 withavidin-HRP: (A) cross-linking of Cys pairs 125/354 or 129/350;and (B) cross-linking of Cys pairs 122/357 or 116/361.

FIGURE 5: Cross-linking of N6/C6 permease and the effect of ligand.Membranes were prepared from cells expressing N6 and C6 withpaired Cys residues at the positions indicated. Cross-linking wascarried out at room temperature for 5 min in the absence or presenceof 10 mM TDG as indicated, and samples were analyzed asdescribed in Figure 3 with avidin-HRP: (A) Cys pairs 119/238;and (B) Cys pairs 119/242.

1718 Biochemistry, Vol. 38, No. 6, 1999 Wu et al.

shown, when paired Cys residues are placed in helices IVand VII, cross-linking is evident only with paired Cysresidues in the middle of the two transmembrane domains(position 119 or 122 in helix IV; position 238, 239, 242, or243 in helix VII). In contradistinction, no cross-linking isobserved when paired Cys residues are placed in either endof the two helices. The results are consistent with thesuggestion that helices IV and VII cross in scissors-likefashion (Figure 6).

To investigate the tertiary contacts between helices IV andXI, we tested cross-linking in N6/C6 permease containingpaired Cys residues in the two transmembrane domains byusing the same approach. When paired Cys residues areplaced at positions near the cytoplasmic ends, high-efficiencycross-linking is observed with Cys pairs 125/354 or 129/350. In contrast, no cross-linking is evident when paired Cysresidues are introduced into the periplasmic halves of helicesIV (position 114, 116, 119, or 122) and XI (position 357,361, or 364), suggesting that the two helices tilt toward eachother in traversing the membrane from the periplasmic tothe cytoplasmic face. Taken together, the results support theconclusion that helix IV is tilted in such a manner that themiddle is close to helix VII, and the cytoplasmic end is closeto helix XI (Figure 6).

Tilting of helices IV and XI (Figure 6) is consistent withprevious observations. It has been shown (18) that helicesII and VII are in close proximity at the periplasmic endsand tilt away from each other toward the cytoplasmic sideof the membrane. Tilting in this manner leaves space between

helices II and VII at the cytoplasmic ends. Moreover, cross-linking of paired Cys residues in helices II and XI demon-strates that the two helices are in close proximity at thecytoplasmic ends, but not the periplasmic ends. Therefore,it is reasonable to suggest that the cytoplasmic end of helixXI tilts into the cytoplasmic interface between helices II andVII (Figure 6). Finally, tilting the cytoplasmic end of helixIV toward the cytoplasmic interface between helices II andVII is consistent with the observations that the cytoplasmicends of helices IV and XI are in close proximity and thathelices IV and VII cross in such a manner that they are inclose proximity in the middle of the membrane.

For the elucidation of the mechanism of sugar/H+symport,dynamic information is clearly required in addition. In thisrespect, ligand-induced conformational changes have beenobserved in many regions of lac permease (18, 19, 28, 30-35). As shown here, the cross-linking efficiency of Cys pairs119/238 or 119/242 is increased in the presence of ligand.Since binding of TDG increases cross-linking efficiency withboth the 6 and 10 Å cross-linkers, it is unlikely that increasedcross-linking reflects a decrease in interhelical distance inthis instance. Consistently, iodine-catalyzed disulfide forma-tion between the two pairs of Cys residues is not altered.Therefore, it is likely that TDG binding increases thereactivity of the Cys residue at position 119 rather thancausing a decrease in distance between helices IV and VII.Nonetheless, the finding suggests that the interface betweenthe two helices is conformationally active.

FIGURE 6: Helix packing model of lactose permease highlighting the tertiary contacts between helices IV, VII, and XI. Twelve transmembranedomains viewed from the cytoplasmic surface of the membrane are displayed as helical ribbons. The six essential residues [Glu126 (helixIV), Arg144 (helix V), Glu269 (helix VIII), Arg302 (helix IX), and His322 and Glu325 (helix X)] are shown as numbered enlarged blueballs. The two Asp-Lys charge pairs [Asp-240 (helix VII)/Lys-319 (helix X) and Asp-237 (helix VII)/Lys-358 (helix XI)] are shown asnumbered enlarged purple balls. Positions that cross-link between helices IV (orange) and VII (green) are highlighted as red balls. Positionsthat cross-link between helices IV and XI (dark red) are highlighted as green balls. Positions that do not exhibit cross-linking are depictedas gray balls. The cytoplasmic end of helix VII tilts away from helices I, II, IV, and V. Tilting of helices I, II, V, VII, VIII, X, and XI isderived from previous studies (18, 19, 28).

Helix Tilting in lac Permease Biochemistry, Vol. 38, No. 6, 19991719

REFERENCES

1. Kaback, H. R. (1976)J. Cell. Physiol. 89, 575-593.2. Kaback, H. R. (1983)J. Membr. Biol. 76, 95-112.3. Kaback, H. R. (1989)HarVey Lect. 83, 77-103.4. Poolman, B., and Konings, W. N. (1993)Biochim. Biophys.

Acta 1183, 5-39.5. Varela, M. F., and Wilson, T. H. (1996)Biochim. Biophys.

Acta 1276, 21-34.6. Viitanen, P., Newman, M. J., Foster, D. L., Wilson, T. H.,

and Kaback, H. R. (1986)Methods Enzymol. 125, 429-452.7. Sahin-To´th, M., Lawrence, M. C., and Kaback, H. R. (1994)

Proc. Natl. Acad. Sci. U.S.A. 91, 5421-5425.8. Kaback, H. R. (1996) inHandbook of Biological Physics:

Transport Processes in Eukaryotic and Prokaryotic Organisms(Konings, W. N., Kaback, H. R., and Lolkema, J. S., Eds.) pp203-227, Elsevier, Amsterdam, The Netherlands.

9. Kaback, H. R., Voss, J., and Wu, J. (1997)Curr. Opin. Struct.Biol. 7, 537-542.

10. Kaback, H. R., and Wu, J. (1997)Q. ReV. Biophys. 30, 333-364.

11. Frillingos, S., Gonzalez, A., and Kaback, H. R. (1997)Biochemistry 36, 14284-14290.

12. Frillingos, S., Sahin-To´th, M., Wu, J., and Kaback, H. R.(1999)FASEB J. 12, 1281-1299.

13. Costello, M. J., Viitanen, P., Carrasco, N., Foster, D. L., andKaback, H. R. (1984)J. Biol. Chem. 259, 15579-15586.

14. Costello, M. J., Escaig, J., Matsushita, K., Viitanen, P. V.,Menick, D. R., and Kaback, H. R. (1987)J. Biol. Chem. 262,17072-17082.

15. Li, J., and Tooth, P. (1987)Biochemistry 26, 4816-4823.16. Zhuang, J., Prive´, G. G., Verner, G. E., Ringler, P., Kaback,

H. R., and Engel, A. (1999)J. Struct. Biol. (in press).17. Wu, J., and Kaback, H. R. (1996)Proc. Natl. Acad. Sci. U.S.A.

93, 14498-14502.18. Wu, J., Hardy, D., and Kaback, H. R. (1998)J. Mol. Biol.

282, 959-967.19. Wu, J., Hardy, D., and Kaback, H. R. (1998)Biochemistry

37, 15785-15790.

20. Bibi, E., and Kaback, H. R. (1990)Proc. Natl. Acad. Sci.U.S.A. 87, 4325-4329.

21. Frillingos, S., Sahin-Toth, M., Persson, B., and Kaback, H.R. (1994)Biochemistry 33, 8074-8081.

22. Dunten, R. L., Sahin-Toth, M., and Kaback, H. R. (1993)Biochemistry 32, 12644-12650.

23. Consler, T. G., Persson, B. L., Jung, H., Zen, K. H., Jung, K.,Prive, G. G., Verner, G. E., and Kaback, H. R. (1993)Proc.Natl. Acad. Sci. U.S.A. 90, 6934-6938.

24. Wu, J., Sun, J., and Kaback, H. R. (1996)Biochemistry 35,5213-5219.

25. Sanger, F., Nicklen, S., and Coulsen, A. R. (1977)Proc. Natl.Acad. Sci. U.S.A. 74, 5463-5467.

26. Carrasco, N., Herzlinger, D., Mitchell, R., DeChiara, S., Danho,W., Gabriel, T. F., and Kaback, H. R. (1984)Proc. Natl. Acad.Sci. U.S.A. 81, 4672-4676.

27. Zen, K., Consler, T. G., and Kaback, H. R. (1995)Biochemistry34, 3430-3437.

28. Wu, J., and Kaback, H. R. (1997)J. Mol. Biol. 270, 285-293.

29. Sun, J., and Kaback, H. R. (1997)Biochemistry 36, 11959-11965.

30. Sun, J., Kemp, C. R., and Kaback, H. R. (1998)Biochemistry37, 8020-8026.

31. Wu, J., Frillingos, S., Voss, J., and Kaback, H. R. (1994)Protein Sci. 3, 2294-2301.

32. Wu, J., Frillingos, S., and Kaback, H. R. (1995)Biochemistry34, 8257-8263.

33. Frillingos, S., Wu, J., Venkatesan, P., and Kaback, H. R. (1997)Biochemistry 36, 6408-6414.

34. Frillingos, S., and Kaback, H. R. (1996)Biochemistry 35,3950-3956.

35. Sahin-To´th, M., and Kaback, H. R. (1993)Protein Sci. 2,1024-1033.

BI982342B

1720 Biochemistry, Vol. 38, No. 6, 1999 Wu et al.